The other 70 percent

What if we only had to be as effective as most cancer drugs?

Imagine going through life giving less than 100 percent and still you're considered a smashing success. Imagine giving 30 percent: 30 percent on every school test and class, 4.0. Every form and application you fill out, you stop a third of the way in and you're still accepted.

Many cancer drugs do just that, yet they get by and are sold as safe, viable treatment options on the US market.

THE PIXANTRONE TRAIN WRECK

Most recently is the case of pixantrone. I just read through the FDA's briefing document concerning the New Drug Application for pixantrone submitted by Cell Therapeutics Inc (CTI) to treat people with relapsed/refractory NHL with two prior failed treatments.

It's pathetic. A train wreck.

On March 22 2010 the FDA's Oncologic Drugs Advisory Committee (ODAC) voted 9-0 to reject the NDA. If they hadn't it would have been an embarrassment.

CTI envisions pixantrone as an option that currently doesn't exist in NHLs: a so-called third-line therapy for people with aggressive NHLs who failed first-line therapy and failed second-line therapy. Right now, NCCN treatment guidelines for these people are grim. They can:

1. Take part in a clinical trial
2. Be given palliative radiation
3. Receive 'supportive care'

So CTI wanted approval from the FDA to market pixantrone as a potential first: a safe, effective third-line therapy for aggressive NHLs. So they pitched the drug and their supporting documentation to the ODAC. Leading the way was their pivotal EXTEND phase III clinical trial (Expanding the reach of anthracyclines with piXanTronE in relapsed or refractory aggressive NHL Disease).

The EXTEND trial was supposed to recruit 320 but only managed 140, which they split evenly into two arms: one received pixantrone, the other received one of several other established chemotherapuetic drugs. Overall response in the pixantrone arm—meaning the number or percentage of patients in this study who responded to pixantrone, whether completely or partially—was 26 of 70 (37%).

Complete response—meaning the number or percentage of patients who showed no signs of cancer at testing—was 8 of 70 (11%).

Furthermore, despite having been touted as an anthracycline without the usual high cardiotoxicity associated with that class of drugs, pixantrone proved more cardiotoxic than is acceptable.

Even more outrageous is that CTI went to the US FDA to get approval to sell pixantrone to American NHL patients, yet they only recruited 8 people from the States. And none of them were among the handful in the study who achieved complete response!

In short, CTI's application was startlingly amateurish and full of gaping holes. The FDA's final decision comes in April; the agency isn't beholden to ODAC's recommendations but they generally follow it.

VELCADE (bortezomib)

This reminds me of another trial …

… the one in 2006 that convinced the FDA to approve Velcade (bortezomib) as a second-line therapy for patients with mantle cell lymphoma. MCL is poorly understood, and there isn't much out there. Millenium Pharmaceuticals already had FDA approval for Velcade for multiple myeloma—like MCL a rare cancer with few treatment options.

So, Millenium recruited 155 patients for their pivotal phase III clinical trial.

The overall response rate was 31%. 48 people. And on average, this response lasted not even ten months before their cancers returned.

However, remove the MCL patients who only had a partial response and count only those with a complete response to Zelcade ...

8%. Twelve patients! The response?

Meanwhile a full 107 patients didn't get a damn thing out of this.

OK that's not entirely true, since a glance at the side effects suggests that Zelcade is much more effective at making cancer patients feel shitty than at making them feel better. After all, while a mere 31 percent enjoyed some anti-cancer benefits, look at how many enjoyed other things:

111 (72%) experienced asthenic conditions
85 (55%) experienced peripheral neuropathies
77 (50%) had constipation
73 (47%) got diarrhea
68 (44%) were nauseaus
60 (39%) had decreased appetites

Half were constipated, half had diarrhea. That makes perfect sense ...

MISAPPROPRIATION OF EMPHASIS

The FDA justifies the approval of these medications because the NHL subtype is rare and not well represented in approved therapeutic agents. Consider Folotyn (pralatrexate), approved for peripheral T-cell lymphoma (PTCL). It performed even worse than those mentioned above, generating some anti-cancer activity in just 27%.

In cases like these, the FDA misappropriates emphasis. They look at the fraction of patients who appeared to get a sliver of benefit and say, 'well, better some receive a few months' benefit by approval than none by denial' but is this what's going on? Sure, when you put a positive spin on it, but when you dispense with that and confront reality, you see that the huge majority of patients just get side effects.

A cohort of 140, or 155, is not large enough (by at least half) to reach a conclusion of any real significance and to extrapolate it across bigger numbers. In the mind of any statistician worth his figures, this invalidates the results. And when complete response is as low as it is in these drugs, you can't even be certain that the drug had anything to do with the response—there simply isn't enough data.

LOW-STANDARDS MAN, or DEATH-CAMP DIRECTOR?

The FDA's Office of Oncology Drug Products (OODP) is something like the Will Rogers of cancer drugs in that they rarely meet one that they don't like. This office doesn't have to follow the recommendations of the ODAC; the decision comes down to Dr. Richard Pazdur, a former practicing oncologist and current director of the OODP. About the difficulties faced by his office, he told the New York Times, "You can’t win in this job. If you approve a drug, they accuse you of lowering standards. And if you don’t approve it, you’re the worst thing since the Nazi death camps and should be killed.”

He's right, I think their standards are too low, especially since there is another option.

COMPASSIONATE USE

Even though it hasn't proven itself to be much better than a chiclet, pixantrone has been available in Europe under the named patient program, given to healthcare professionals who request it for the treatment of individual patients. The FDA has a similar structure in place, generically known as compassionate use. That's where these substandard drugs belong: They shouldn't be presented to the cancer community as being viable treatment options when they perform as poorly as they do, when they cost as much as they do, when they exploit as much hope as they do.

It's not fair to the other 70 percent.